Patterns of Gene Expression in Early Developing Complex Leaves

Ciera Martinez
October 18, 2014



cell differentiation



cell differentiation



cell differentiation



Pattern 1: Tip to base wave of differentiation.
Cell differentation occurring first at the tip.



cell differentiation



Pattern 1: Tip to base wave of differentiation.
Cell differentation occurring first at the tip.

Pattern 2: Midrib to Marginal blastozone or marginal meristem
(defined histologically as having dense cells and maintenance of high rates of cell division)



cell differentiation



Pattern 1: Tip to base wave of differentiation.
Cell differentation occurring first at the tip.

Pattern 2: Midrib to Marginal blastozone or marginal meristem
(defined histologically as having dense cells and maintenance of high rates of cell division)

Main Question



What are the genes expression trends observed in early complex leaf development that would explain differentiation patterning in the leaf?

Approach

cell differentiation

Isolate tissue

1. longitudinal axis (tip, mid, base)

2. margins compared with all other tissue (rachis and midvein) regions, to perform gene expression analysis.

Laser Capture Microdissection

Differential Expression (DE) & GO enrichment

cell differentiation

DE & GO enrichment

cell differentiation



Conclusion

The regions further along in development will have up-regulation of GO categories associated with differentiated tissue.

Photosynthetic Activity

cell differentiation

Photosynthetic Activity

cell differentiation

It is unknown when and where photosynthetic activity is first evident in early developing leaves.

Photosynthetic Activity

cell differentiation

Approach: Chlorophyll a/b binding protein (CAB)::GUS
localization reflects photosynthetic activity.

Photosynthetic Activity

cell differentiation

Approach: Chlorophyll a/b binding protein (CAB)::GUS
localization reflects photosynthetic activity.

Ubiquitous in mature leaves.

Photosynthetic Activity

cell differentiation

Approach: Chlorophyll a/b binding protein (CAB)::GUS
localization reflects photosynthetic activity.

Ubiquitous in mature leaves.

Where does CAB localize early in leaf development?

Photosynthetic Activity

cell differentiation

Photosynthetic Activity


Conclusions 1: Early in leaf development (P4 & P5), the rachis and midviein show CAB activity, suggesting these regions are first to start specialized processes such as photosynthesis.

Conclustion 2: The LCM approach for determining gene expression patterns is capable of predicting verifiable expression patterns.

Co-expression Analysis

cell differentiation

Question 1: Does clustering give similar GO enrichment results to DE analysis?

Question 2: Are there patterns of gene expression that explain margin/rachis identity?

Question 3: Can we get to smaller subset of genes that may explain differentiation patterning during leaf development?

Principle Component Analysis

plot of chunk unnamed-chunk-2

PC1 (29.2%)

PC2 (20.8%)

plot of chunk unnamed-chunk-3
PC3 (20.3%)

PC4 (15.6%)

Self Organizing Maps





Groups objects based on categories that define them.

Self Organizing Maps

map

Self Organizing Maps: Limit to six clusters

map

plot of chunk unnamed-chunk-4

Clusters help define PCA clustering

plot of chunk unnamed-chunk-5

plot of chunk unnamed-chunk-6





Question 1: What genes define these clusters? Does clustering give similar GO enrichment results to DE analysis?

Cluster 2

plot of chunk unnamed-chunk-7

Cluster 2 - Genes that are up-regulated in top midvein

aother

plot of chunk unnamed-chunk-8

Cluster 2 - Photosynthetic GO categories

aother

Question 1: Does clustering give similar GO enrichment results to DE analysis?

GO Categories

  • heme binding
  • oxygen binding
  • photosynthesis, light harvesting
  • apoptotic process

Yes. Recapitulates DE expression patterns in tissue specific regions.

cell differentiation



Question 2: What are the specifc genes that contribute to marginal idenity?

Approach: Make a larger self organizing map allowing clusters that can specify regulated genes across multiple tissue types.

Self Organizing Maps - Allow 36 clusters

plot of chunk unnamed-chunk-9

map

Cluster 17 Up-regulated in rachis compared to margin

map

plot of chunk unnamed-chunk-10

Cluster 17: Growth Genes

Auxin Response 6 - auxin response via expression of auxin regulated genes

gibberellin 2-oxidase - responsive to cytokinin and KNOX activities

ARGONAUTE7 - required for mediolateral expansion in maize (Douglas et al, 2010)

REDUCED STEM BRANCHING 6 - MADS-box transcription factor, mutant flower margins expanded (TAIR mutant lines)

AP2/B3 domain transcription factor - may function as a negative growth regulator

R2R3-MYB TF factor gene - MYB gene involved in cell fate idenity & Lateral Meristem Initiation (Muller et al., 2005)

EMBRYO DEFECTIVE - mutant with enlarged SAM (Cushing et al. 2005)

Cluster 35

map

plot of chunk unnamed-chunk-11

Cluster 35 - Photosynthesis and Vascular Differentiation

Solyc00g277510 - Encodes chlorophyll binding protein D1, a part of the photosystem II reaction center core

Solyc02g071000 - Subunit of light-harvesting complex II (LHCII),which absorbs light and transfers energy to the photosynthetic reaction center.

Solyc05g013570 - phototropic-responsive NPH3 family protein

Solyc05g041230 - chloroplast gene encoding a CP43 subunit of the photosystem II reaction center.

Solyc08g066500 - Member of the class III HD-ZIP protein family. Critical for vascular development.

Solyc08g067330 - Encodes lhcb1.1 a component of the LHCIIb light harvesting complex associated with photosystem II.

trifoliate (tf-2) mutant

trifoliate


The TF-2 gene is required to maintain morphogenetic potential (PNAS, Naz et al., 2013).

trifoliate mutant

trifoliate

Marginal Blastozone in trifoliate cannot make new leaflets at P4 stage.

Comparisons with wildtype will help isolate genes involved in regulation of morphogenetic potential.

Super Self Organized Maps (superSOM)



somDiagram



Super SOM: clusters have dimensionality and a separate identity associated with genotype data set, but ultimately, data must be assigned to the same cluster.

SuperSOM - Cluster 12

leaf

plot of chunk unnamed-chunk-13

SuperSOM - Cluster 12

plot of chunk unnamed-chunk-14

plot of chunk unnamed-chunk-15

SuperSOM - Cluster 12

Auxin Regulation and Transport

Solyc12g006340.1.1 (ARF8):Encodes a member of the auxin response factor family.

Solyc10g076790.1.1 (AUX1): Encodes an auxin influx transporter.

Solyc03g118740.2.1 (PIN1): Auxin efflux.

SuperSOM - Cluster 12

Developmental Transciption Factors

Solyc09g065820.2.1: DNA binding / transcription factor; cell differentiation.

Solyc09g010780.2.1: Involved in leaf development. Knockout mutants have abnormally shaped leaves.

Solyc02g080260.2.1: Encodes a homeodomain protein that is expressed in the LI layer of the vegetative

Solyc07g018290.2.1 (PLETHERA 1): expressed in young tissues and may specify meristematic or division-competent states. (Wilson et al., 2005).

Solyc06g075850.1.1: Histone H4 : cytochrome P450 monooxygenase. Expressed in cotyledons and leaves

Solyc04g009950.2.1: R2R3- type MYB- encoding genes

SuperSOM - Cluster 28

leaf

plot of chunk unnamed-chunk-16

SuperSOM - Cluster 28

plot of chunk unnamed-chunk-17

plot of chunk unnamed-chunk-18

Conclusions

  1. There are GO categories associated with different levels of cell differentiation.

  2. There are groups of genes that define marginal blastozone along longitudinal axis, which include both known regulators and unknown genes.

Future

  • Combine all these interesting developmental clusters.

  • Combining both data from this study and other RNAseq leaf data. Network analysis to identify major hubs of developmental regulation.

  • Functional analysis.

Acknowlegements



Thanks

Sinha Lab

*Brad Townsley

Harada Lab



Funding

NSF GRFP Fellowship

Elsie Taylor Stocking Fellowship

Walter R. and Roselinde H. Russell Fellowship

Collaborative Coding

On Github:

-Scripts

-Normalized Data

-Tools for SOM cluster visualization

@iamciera on Github.

Coding Collective

Collaboratively learn to program.

All programing languages.

All skill levels.

Learn faster, code together.

4-6pm (in LSA 2002)

github.com/theCodingCollective

leaf